Well wall stability prediction and optimization system based on real-time monitoring data

Abstract

A well wall stability prediction and optimization system based on real-time monitoring data relates to the technical field of oil drilling and comprises the following steps: acquiring real-time multi-source sensing data of a target well wall, marking an acquisition moment and setting an acquisition period; constructing a three-dimensional geometric model of the well wall and a stratum geometric model; setting a transition area, and performing grid division on the three-dimensional geometric model, the stratum geometric model and the transition area; acquiring scene features of each tetrahedral mesh in the three-dimensional geometric model and the stratum geometric model; performing real-time monitoring on the scene features of each tetrahedral mesh; and carrying out stability evaluation on each tetrahedral mesh, and carrying out mesh density refinement operation or prediction operation based on an evaluation result. According to the method, the accuracy and reliability of well wall stability prediction and optimization are improved, and the efficiency and safety of drilling engineering are improved.

Description

Technical Field

[0001] The invention relates to the technical field of oil drilling, in particular to a wellbore stability prediction and optimization system based on real-time monitoring data. Background Art

[0002] The prior art CN113221347A "A method, device and equipment for optimizing wellbore stability drilling" includes: using seismic data and logging data of adjacent wells to establish a nonlinear mapping relationship between seismic parameters and logging parameters; establishing a formation pore pressure prediction model based on the logging data and formation pore pressure data of adjacent wells, and using the logging data obtained by inversion as input to obtain formation pore pressure data; establishing a formation lithology, rock mechanical properties and ground stress prediction model based on the logging data and core data of adjacent wells, and inputting the logging data obtained by inversion to obtain the corresponding formation lithology, rock mechanical properties and ground stress; substituting the rock mechanical parameters into the mechanism model to calculate the formation collapse pressure and formation fracture pressure respectively; based on the formation pressure, lithology, drilling fluid and drilling design parameters, optimizing the machine learning algorithm, and establishing a pre-drilling intelligent prediction model for wellbore stability.

[0003] The prior art CN111691873B "Wellbore stability value calculation method and system for wellbore stability prediction" includes: determining the associated influencing parameters affecting the wellbore stability value based on historical statistical data; and constructing a calculation model for calculating the wellbore stability value based on the association analysis method based on the associated influencing parameters; for the well to be analyzed, collecting the measured data of the well to be analyzed; and calculating the wellbore stability value of the well to be analyzed based on the measured data and the calculation model.

[0004] In the process of oil drilling, wellbore stability is an important factor affecting drilling safety and efficiency. Wellbore instability may lead to serious accidents such as well leakage and blowout, which not only increases drilling costs but also may cause environmental pollution. Traditional wellbore stability assessment methods mainly rely on geological data and empirical formulas, which lack real-time and accuracy. Therefore, it is of great significance to develop a wellbore stability prediction and optimization system based on real-time monitoring data. Summary of the invention

[0005] In order to solve the above technical problems, the purpose of the present invention is to provide a wellbore stability prediction and optimization system based on real-time monitoring data, including a monitoring center, wherein the monitoring center is communicatively connected with a data acquisition module, a model building module, a grid division module, a data processing module, a real-time monitoring module and an optimization prediction module;

[0006] The data acquisition module is used to collect real-time multi-source sensor data of the target well wall, mark the acquisition time, and set the acquisition cycle;

[0007] The model building module is used to build a three-dimensional geometric model of the well wall and a stratum geometric model;

[0008] The grid division module is used to set the transition area and perform grid division on the three-dimensional geometric model, the stratum geometric model and the transition area;

[0009] The data processing module is used to obtain scene features of each tetrahedral grid in the three-dimensional geometric model and the stratum geometric model;

[0010] The real-time monitoring module is used to monitor the scene features of each tetrahedral grid in real time;

[0011] The optimization prediction module is used to evaluate the stability of each tetrahedral mesh and perform mesh density refinement or prediction operations based on the evaluation results.

[0012] Furthermore, the process of constructing the three-dimensional geometric model of the wellbore and the formation geometric model by the model building module includes:

[0013] The drilling trajectory, wellbore parameters, geological stratification, stratigraphic structure, lithological characteristics, load attributes and drilling parameters are acquired based on real-time multi-source sensing data. A three-dimensional geometric model of the wellbore is constructed based on the load, drilling trajectory and wellbore parameters. A stratigraphic geometric model of the wellbore is constructed based on the load, geological stratification, geological structure and rock characteristics.

[0014] Furthermore, the process of meshing the three-dimensional geometric model and the stratum geometric model by the meshing module includes:

[0015] Obtaining a well wall grid density of a well wall region in the three-dimensional geometric model according to the well wall parameters in the three-dimensional geometric model, and performing grid division on the well wall region in the three-dimensional geometric model according to the well wall grid density, dividing the well wall region into a plurality of tetrahedral grids of the same specification;

[0016] Obtaining geological structures and rock characteristics in the stratigraphic geometry model Obtain stratigraphic grid density of different stratigraphic regions in the stratigraphic geometry model, and meshing different stratigraphic regions in the stratigraphic geometry model according to the stratigraphic grid density into a number of tetrahedral grids of the same specification.

[0017] Furthermore, the grid division module sets a transition area, and the process of grid division of the transition area includes:

[0018] A wellbore transition diameter is preset, and according to the wellbore transition diameter and the wellbore outer diameter in the three-dimensional geometric model, the formation area between the wellbore outer diameter and the wellbore transition diameter in the formation geometric model is marked as a transition area. At the same time, several threshold outer diameters are selected between the wellbore outer diameter and the wellbore transition diameter, and the transition area is divided into different transition sub-intervals. The grid density of each transition sub-interval is set according to the threshold outer diameter of each transition sub-interval, the formation grid density, the wellbore outer diameter and the wellbore grid density.

[0019] Furthermore, the process of the data processing module acquiring the scene features of each tetrahedral mesh in the three-dimensional geometric model and the stratum geometric model includes:

[0020] Obtain the coordinate positions, load attributes and drilling parameters of the vertices of the tetrahedral mesh in the three-dimensional geometric model and the formation geometric model, obtain the coordinate position of the geometric center of the tetrahedral mesh in the three-dimensional geometric model, obtain the Euclidean distances between the vertices and the geometric centers of other tetrahedral meshes in the three-dimensional geometric model and the formation geometric model, and construct scene features of the tetrahedral mesh based on the Euclidean distances, load attributes and drilling parameters of the vertices of other tetrahedral meshes and the load attributes and drilling parameters of the tetrahedral mesh.

[0021] Furthermore, the process of the real-time monitoring module performing real-time monitoring on the scene features of each tetrahedral grid includes:

[0022] Obtain the scene features of each tetrahedral mesh in the three-dimensional geometric model and the formation geometric model in the current acquisition cycle, preset the threshold intervals corresponding to each type of indicator in the load attributes of the vertices of each tetrahedral mesh and the threshold intervals corresponding to each type of indicator in the drilling parameters, compare the load attributes and drilling parameters in the scene features of each tetrahedral mesh in the current acquisition cycle with the corresponding threshold intervals, and if a certain type of indicator in the load attribute is not located in the corresponding threshold interval or a certain type of indicator in the drilling parameter is not located in the corresponding threshold interval, an abnormal alarm signal is generated, and the monitoring center arranges relevant personnel to perform detection and analysis according to the abnormal alarm signal, obtains the corresponding drilling parameters, presets a control database, and matches the scene features of the tetrahedral mesh with the drilling parameters and stores them in the control database;

[0023] If each type of index in the load attribute and each type of index in the drilling parameter are within the corresponding threshold range, the wellbore stability optimization operation is performed.

[0024] Furthermore, the optimization prediction module evaluates the stability of each tetrahedral mesh and performs a mesh density refinement operation or a prediction operation based on the evaluation result, including:

[0025] Obtaining scene features of each tetrahedral grid in a three-dimensional geometric model and a formation geometric model in several historical acquisition cycles, performing statistical analysis on load attributes and drilling parameters in the scene features, obtaining the cumulative number of times each type of indicator in the load attributes and drilling parameters is not located in a corresponding threshold interval, and obtaining weight coefficients of each type of indicator in the load attributes and drilling parameters according to the cumulative number;

[0026] Analyze the load attributes and drilling parameters in the scene features of each tetrahedral mesh in the three-dimensional geometric model and the formation geometric model in the current acquisition cycle, obtain the fluctuation coefficients of various types of indicators in the load attributes and drilling parameters, use the fluctuation coefficients of various types of indicators in the load attributes and drilling parameters of each tetrahedral mesh as evaluation indicators, preset the stability level, and obtain the membership matrix of each tetrahedral mesh to the stability level through fuzzy comprehensive evaluation;

[0027] The stability level of each tetrahedral mesh is obtained according to the membership matrix and the weight coefficient of each type of indicator. The mesh density is refined for the tetrahedral meshes whose stability level is less than the preset stability level threshold, and then the prediction operation is performed on each tetrahedral mesh.

[0028] Furthermore, the process of performing a mesh density refinement operation includes:

[0029] A mesh density comparison table is constructed, wherein the mesh density comparison table includes mesh densities corresponding to different stability coefficients, the mesh density of the tetrahedral mesh is obtained according to the mesh density comparison table and the stability coefficient of the tetrahedral mesh, and the tetrahedral mesh is divided into a number of new tetrahedral meshes of the same specification according to the mesh density.

[0030] Furthermore, the process of performing prediction operations on each tetrahedral mesh includes:

[0031] A wellbore prediction model is constructed based on deep learning, and scene features of each tetrahedral grid in a three-dimensional geometric model and a formation geometric model in several historical acquisition cycles are used as a training set and a test set. The training set is input into the wellbore prediction model for training until the loss function training is stable, and the model parameters are saved. The wellbore prediction model is tested by the test set until it meets the preset requirements, and the wellbore prediction model is output;

[0032] According to the prediction scene features of each tetrahedral mesh output by the wellbore prediction model, the load attributes and drilling parameters in the prediction scene features of each tetrahedral mesh are compared with the corresponding threshold intervals. If a certain type of indicator in the load attribute of the tetrahedral mesh is not located in the corresponding threshold interval or a certain type of indicator in the drilling parameter is not located in the corresponding threshold interval, the prediction scene features are input into the control database for data retrieval, and scene features whose similarity meets the preset standard are obtained in the control database, and the drilling parameters matching the scene features are obtained, and the drilling parameters are used as new drilling parameters of the tetrahedral mesh.

[0033] Compared with the prior art, the present invention provides a wellbore stability prediction and optimization system based on real-time monitoring data. By integrating multiple functional modules, it realizes comprehensive monitoring, accurate prediction and optimization adjustment of wellbore stability. The following are the detailed beneficial effects of the system:

[0034] 1. The system can collect real-time data from a variety of sensors, including drilling trajectory, wellbore parameters, geological stratification, stratigraphic structure, lithological characteristics, load attributes and drilling parameters, ensuring the comprehensiveness and accuracy of the data. The system can monitor the scene characteristics of each tetrahedral grid in real time, detect abnormal situations in a timely manner, and improve the real-time and response speed of monitoring.

[0035] 2. By constructing a three-dimensional geometric model of the well wall and formation, the system can simulate the geometric structure of the well wall and formation more accurately, improve the accuracy and adaptability of the model, and the system can perform grid division according to the well wall parameters and formation characteristics to ensure the rationality and uniformity of the grid, thereby improving the resolution and calculation accuracy of the model.

[0036] 3. The system can set the transition area and divide multiple transition sub-intervals between the outer diameter of the well wall and the transition diameter of the well wall to ensure a smooth transition of the grid density and avoid computational instability caused by sudden changes in the grid size. The system can refine the grid density according to the stability coefficient of the tetrahedral grid to ensure that smaller grids are used in areas where high precision is required, thereby improving the accuracy and efficiency of the calculation. At the same time, the system can perform local refinement in areas where high precision is required to avoid the waste of computational resources caused by global refinement, thereby improving computational efficiency and resource utilization.

[0037] 4. Through the fuzzy comprehensive evaluation method, the system can comprehensively consider multiple factors and obtain the stability level of each tetrahedral grid, thereby improving the accuracy and reliability of the evaluation. The system builds a wellbore prediction model based on deep learning, which can be trained and tested according to historical data, and output the predicted scenario characteristics of each tetrahedral grid, thereby improving the accuracy and reliability of the prediction. The system can match the predicted scenario characteristics with the historical data in the control database, obtain the optimal drilling parameters, and realize intelligent optimization and adjustment.

[0038] Through the above beneficial effects, the present invention not only improves the accuracy and reliability of wellbore stability prediction and optimization, but also improves the efficiency and safety of drilling projects, and has important practical application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 This is a schematic diagram of a wellbore stability prediction and optimization system based on real-time monitoring data according to an embodiment of the present application. DETAILED DESCRIPTION

[0040] The following is a clear and complete description of the technical solutions in the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0041] like Figure 1 As shown, a wellbore stability prediction and optimization system based on real-time monitoring data includes a monitoring center, and the monitoring center is communicatively connected with a data acquisition module, a model building module, a grid division module, a data processing module, a real-time monitoring module and an optimization prediction module;

[0042] The data acquisition module is used to collect real-time multi-source sensor data of the target well wall, mark the acquisition time, and set the acquisition cycle;

[0043] The model building module is used to build a three-dimensional geometric model of the well wall and a stratum geometric model;

[0044] The grid division module is used to set the transition area and perform grid division on the three-dimensional geometric model, the stratum geometric model and the transition area;

[0045] The data processing module is used to obtain scene features of each tetrahedral grid in the three-dimensional geometric model and the stratum geometric model;

[0046] The real-time monitoring module is used to monitor the scene features of each tetrahedral grid in real time;

[0047] The optimization prediction module is used to evaluate the stability of each tetrahedral mesh and perform mesh density refinement or prediction operations based on the evaluation results.

[0048] It should be further explained that, in a specific implementation process, the process of the model building module building the three-dimensional geometric model of the well wall and the formation geometric model includes:

[0049] Based on real-time multi-source sensor data, the drilling trajectory, wellbore parameters (including wellbore thickness, borehole diameter and wellbore outer diameter), geological stratification (including the thickness and position of different rock layers), formation structure (such as faults, cracks and other structures in the formation), lithological characteristics (such as elastic modulus, Poisson's ratio, compressive strength, etc. of rock), load attributes (such as stress, strain, formation pressure, drilling fluid pressure and drill tool load, etc.) and drilling parameters (such as drilling pressure, rotation speed, drilling fluid properties, etc.) are obtained. A three-dimensional geometric model of the wellbore is constructed according to the load, drilling trajectory and wellbore parameters, and a formation geometric model of the wellbore is constructed according to the load, geological stratification, geological structure and rock characteristics.

[0050] It should be further explained that, in a specific implementation process, the process of meshing the three-dimensional geometric model and the stratum geometric model by the meshing module includes:

[0051] Obtaining the well wall grid density of the well wall area in the three-dimensional geometric model according to the well wall parameters in the three-dimensional geometric model, meshing the well wall area in the three-dimensional geometric model according to the well wall grid density, and dividing it into a number of tetrahedral grids of the same specification, wherein a tetrahedral unit is a tetrahedral geometric body with four vertices (nodes) and four faces, each face is a triangle, and the vertex of each tetrahedral unit is called a node, and each node has a coordinate position and a load attribute. The tetrahedral grid can adapt well to complex geometric shapes and is suitable for processing irregular or curved geometric models, and the tetrahedral grid can be easily locally encrypted to improve the calculation accuracy of a specific area. Compared with the hexahedral grid, the generation and processing of the tetrahedral grid is usually simpler and has a higher calculation efficiency, which provides convenience for subsequent calculations;

[0052] Obtaining geological structures and rock characteristics in the stratigraphic geometry model Obtain stratigraphic grid density of different stratigraphic regions in the stratigraphic geometry model, and meshing different stratigraphic regions in the stratigraphic geometry model according to the stratigraphic grid density into a number of tetrahedral grids of the same specification.

[0053] It should be further explained that, in the specific implementation process, the grid division module sets the transition area, and the process of grid division of the transition area includes:

[0054] A wellbore transition diameter is preset, and according to the wellbore transition diameter and the wellbore outer diameter in the three-dimensional geometric model, the formation area between the wellbore outer diameter and the wellbore transition diameter in the formation geometric model is marked as a transition area. In this transition area, the grid size will gradually transition from a fine grid near the wellbore to a coarser grid far away from the formation. At the same time, several threshold outer diameters are selected between the wellbore outer diameter and the wellbore transition diameter, and the transition area is divided into different transition sub-intervals. The grid density of each transition sub-interval is set according to the threshold outer diameter, formation grid density, wellbore outer diameter and wellbore grid density of each transition sub-interval.

[0055] It should be further explained that, in the specific implementation process, according to the threshold outer diameter of each transition sub-interval, the formation grid density, the wellbore outer diameter and the wellbore grid density, the specific process of setting the grid density of each transition sub-interval is as follows:

[0056] According to the formation grid density of the transition sub-interval in each formation area, the outer diameter of the well wall, the well wall grid density of the well wall area at the same depth of the transition sub-interval in each formation area, and the threshold outer diameter of the transition sub-interval in each formation area, the grid density of the transition sub-interval in each formation area is obtained, and the calculation formula is:

[0057] ;

[0058] in, represents the grid density of transition subinterval i, represents the well wall grid density, represents the formation grid density, represents the exponential, k is the exponential decay constant, represents the threshold outer diameter of transition subinterval i, Represents the outer diameter of the well wall. The present invention uses the threshold outer diameter with a larger value of the transition sub-interval as the threshold outer diameter of the transition sub-interval by default. The above formula is a formula closest to the actual situation obtained by software simulation through collecting a large amount of data. The preset parameters and preset thresholds in the formula are set by technicians in this field according to actual conditions or obtained by simulating a large amount of data.

[0059] It should be further explained that, in a specific implementation process, the process of the data processing module acquiring the scene features of each tetrahedral mesh in the three-dimensional geometric model and the stratum geometric model includes:

[0060] Obtain the coordinate positions, load attributes and drilling parameters of the vertices of the tetrahedral mesh in the three-dimensional geometric model and the formation geometric model, the drilling parameters of the vertices of the tetrahedral mesh in the formation geometric model are zero by default, obtain the coordinate position of the geometric center of the tetrahedral mesh in the three-dimensional geometric model, obtain the Euclidean distance between the coordinate positions of the vertices of other tetrahedral meshes in the three-dimensional geometric model and the formation geometric model and the coordinate position of the geometric center, and construct the scene characteristics of the tetrahedral mesh according to the Euclidean distance, load attributes and drilling parameters of the vertices of other tetrahedral meshes and the load attributes and drilling parameters of the tetrahedral mesh.

[0061] It should be further explained that, in the specific implementation process, the process of the real-time monitoring module monitoring the scene features of each tetrahedral grid in real time includes:

[0062] Obtain the scene features of each tetrahedral mesh in the three-dimensional geometric model and the formation geometric model in the current acquisition cycle, preset the threshold intervals corresponding to each type of indicator in the load attributes of the vertices of each tetrahedral mesh and the threshold intervals corresponding to each type of indicator in the drilling parameters, compare the load attributes and drilling parameters in the scene features of each tetrahedral mesh in the current acquisition cycle with the corresponding threshold intervals, and if a certain type of indicator in the load attribute is not located in the corresponding threshold interval or a certain type of indicator in the drilling parameter is not located in the corresponding threshold interval, an abnormal alarm signal is generated, and the monitoring center arranges relevant personnel to perform detection and analysis according to the abnormal alarm signal, obtains the corresponding drilling parameters, presets a control database, and matches the scene features of the tetrahedral mesh with the drilling parameters and stores them in the control database;

[0063] If each type of index in the load attribute and each type of index in the drilling parameter are within the corresponding threshold range, the wellbore stability optimization operation is performed.

[0064] It should be further explained that, in the specific implementation process, the optimization prediction module evaluates the stability of each tetrahedral mesh, and performs a mesh density refinement operation or a prediction operation based on the evaluation result, including:

[0065] Obtaining scene features of each tetrahedral grid in a three-dimensional geometric model and a formation geometric model in several historical acquisition cycles, performing statistical analysis on load attributes and drilling parameters in the scene features, obtaining the cumulative number of times each type of indicator in the load attributes and drilling parameters is not located in a corresponding threshold interval, and obtaining weight coefficients of each type of indicator in the load attributes and drilling parameters according to the cumulative number;

[0066] Analyze the load attributes and drilling parameters in the scene features of each tetrahedral mesh in the three-dimensional geometric model and the formation geometric model in the current acquisition cycle, obtain the fluctuation coefficients of various types of indicators in the load attributes and drilling parameters, use the fluctuation coefficients of various types of indicators in the load attributes and drilling parameters of each tetrahedral mesh as evaluation indicators, preset the stability level, and obtain the membership matrix of each tetrahedral mesh to the stability level through fuzzy comprehensive evaluation;

[0067] The stability level of each tetrahedral mesh is obtained according to the membership matrix and the weight coefficient of each type of indicator. The mesh density is refined for the tetrahedral meshes whose stability level is less than the preset stability level threshold, and then the prediction operation is performed on each tetrahedral mesh.

[0068] It should be further explained that, in the specific implementation process, the process of obtaining the stability level of each tetrahedral mesh according to the membership matrix and the index weight matrix includes:

[0069] An indicator weight matrix of the evaluation indicator is constructed according to the weight coefficient of each type of indicator, and the indicator weight matrix of the evaluation indicator and the membership matrix are fused through a formula to obtain a fuzzy comprehensive evaluation matrix of the evaluation indicator, and the membership of each tetrahedral mesh for different stability levels is obtained according to the fuzzy comprehensive evaluation matrix, and the stability level with the highest membership corresponding to each tetrahedral mesh is screened out, and the stability level with the highest membership corresponding to each tetrahedral mesh is used as the stability level of each tetrahedral mesh;

[0070] Wherein, the formula is:

[0071] ;

[0072] in, is the fuzzy comprehensive evaluation matrix of the evaluation index, is the indicator weight matrix of the evaluation indicator, is the membership matrix, " represents the multiplication of the weight matrix of the evaluation index and the elements at the corresponding positions of the membership matrix, It is a weighted parameter for controlling the balance between the weight matrix and the membership matrix in the fuzzy comprehensive evaluation matrix of the evaluation index.

[0073] It should be further explained that, in the specific implementation process, the process of refining the grid density includes:

[0074] A mesh density comparison table is constructed, wherein the mesh density comparison table includes mesh densities corresponding to different stability coefficients, the mesh density of the tetrahedral mesh is obtained according to the mesh density comparison table and the stability coefficient of the tetrahedral mesh, and the tetrahedral mesh is divided into a number of new tetrahedral meshes of the same specification according to the mesh density.

[0075] Using smaller grids in unstable areas (such as near the wellbore wall, faults, cracks, etc.) can more accurately capture the local stress and strain distribution. Smaller grids can improve the robustness of the wellbore prediction model and reduce the error in the output results of the wellbore prediction model caused by grid quality problems.

[0076] It should be further explained that, in the specific implementation process, the process of performing prediction operations on each tetrahedral mesh includes:

[0077] A wellbore prediction model is constructed based on deep learning, and scene features of each tetrahedral grid in a three-dimensional geometric model and a formation geometric model in several historical acquisition cycles are used as a training set and a test set. The training set is input into the wellbore prediction model for training until the loss function training is stable, and the model parameters are saved. The wellbore prediction model is tested by the test set until it meets the preset requirements, and the wellbore prediction model is output;

[0078] According to the prediction scene features of each tetrahedral mesh output by the wellbore prediction model, the load attributes and drilling parameters in the prediction scene features of each tetrahedral mesh are compared with the corresponding threshold intervals. If a certain type of indicator in the load attribute of the tetrahedral mesh is not located in the corresponding threshold interval or a certain type of indicator in the drilling parameter is not located in the corresponding threshold interval, the prediction scene features are input into the control database for data retrieval, and scene features whose similarity meets the preset standard are obtained in the control database, and the drilling parameters matching the scene features are obtained, and the drilling parameters are used as new drilling parameters of the tetrahedral mesh.

[0079] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. Wellbore stability prediction and optimization system based on real-time monitoring data, characterized by: It includes a monitoring center, which is communicatively connected with a data acquisition module, a model building module, a grid division module, a data processing module, a real-time monitoring module and an optimization prediction module; The data acquisition module is used to collect real-time multi-source sensor data of the target well wall and mark the acquisition time and set the acquisition cycle; The model building module is used to build a three-dimensional geometric model of the well wall and a stratum geometric model; The grid division module is used to set the transition area and perform grid division on the three-dimensional geometric model, the stratum geometric model and the transition area; The data processing module is used to obtain scene features of each tetrahedral grid in the three-dimensional geometric model and the stratum geometric model; The real-time monitoring module is used to monitor the scene features of each tetrahedral grid in real time; The optimization prediction module is used to evaluate the stability of each tetrahedral mesh and perform mesh density refinement or prediction operations based on the evaluation results.

2. The wellbore stability prediction and optimization system based on real-time monitoring data according to claim 1 is characterized in that: The process of constructing the three-dimensional geometric model of the well wall and the formation geometric model by the model construction module includes: The drilling trajectory, wellbore parameters, geological stratification, stratigraphic structure, lithological characteristics, load attributes and drilling parameters are acquired based on real-time multi-source sensing data. A three-dimensional geometric model of the wellbore is constructed based on the load, drilling trajectory and wellbore parameters. A stratigraphic geometric model of the wellbore is constructed based on the load, geological stratification, geological structure and rock characteristics.

3. The wellbore stability prediction and optimization system based on real-time monitoring data according to claim 2 is characterized in that: The process of meshing the three-dimensional geometric model and the stratum geometric model by the meshing module includes: Obtaining a well wall grid density of a well wall region in the three-dimensional geometric model according to the well wall parameters in the three-dimensional geometric model, and performing grid division on the well wall region in the three-dimensional geometric model according to the well wall grid density, dividing the well wall region into a plurality of tetrahedral grids of the same specification; Obtaining geological structures and rock characteristics in the stratigraphic geometry model Obtain stratigraphic grid density of different stratigraphic regions in the stratigraphic geometry model, and meshing different stratigraphic regions in the stratigraphic geometry model according to the stratigraphic grid density into a number of tetrahedral grids of the same specification.

4. The wellbore stability prediction and optimization system based on real-time monitoring data according to claim 3 is characterized in that: The grid division module sets a transition area, and the process of grid division of the transition area includes: A wellbore transition diameter is preset, and according to the wellbore transition diameter and the wellbore outer diameter in the three-dimensional geometric model, the formation area between the wellbore outer diameter and the wellbore transition diameter in the formation geometric model is marked as a transition area. At the same time, several threshold outer diameters are selected between the wellbore outer diameter and the wellbore transition diameter, and the transition area is divided into different transition sub-intervals. The grid density of each transition sub-interval is set according to the threshold outer diameter of each transition sub-interval, the formation grid density, the wellbore outer diameter and the wellbore grid density.

5. The wellbore stability prediction and optimization system based on real-time monitoring data according to claim 4 is characterized in that: The process of the data processing module acquiring the scene features of each tetrahedral mesh in the three-dimensional geometric model and the stratum geometric model includes: Obtain the coordinate positions, load attributes and drilling parameters of the vertices of the tetrahedral mesh in the three-dimensional geometric model and the formation geometric model, obtain the coordinate position of the geometric center of the tetrahedral mesh in the three-dimensional geometric model, obtain the Euclidean distances between the vertices and the geometric centers of other tetrahedral meshes in the three-dimensional geometric model and the formation geometric model, and construct scene features of the tetrahedral mesh based on the Euclidean distances, load attributes and drilling parameters of the vertices of other tetrahedral meshes and the load attributes and drilling parameters of the tetrahedral mesh.

6. The wellbore stability prediction and optimization system based on real-time monitoring data according to claim 5 is characterized in that: The process of the real-time monitoring module performing real-time monitoring on the scene features of each tetrahedral grid includes: Obtain the scene features of each tetrahedral mesh in the three-dimensional geometric model and the formation geometric model in the current acquisition cycle, preset the threshold intervals corresponding to each type of indicator in the load attributes of the vertices of each tetrahedral mesh and the threshold intervals corresponding to each type of indicator in the drilling parameters, compare the load attributes and drilling parameters in the scene features of each tetrahedral mesh in the current acquisition cycle with the corresponding threshold intervals, and if a certain type of indicator in the load attribute is not located in the corresponding threshold interval or a certain type of indicator in the drilling parameter is not located in the corresponding threshold interval, an abnormal alarm signal is generated, and the monitoring center arranges relevant personnel to perform detection and analysis according to the abnormal alarm signal, obtains the corresponding drilling parameters, presets a control database, and matches the scene features of the tetrahedral mesh with the drilling parameters and stores them in the control database; If each type of index in the load attribute and each type of index in the drilling parameter are within the corresponding threshold range, the wellbore stability optimization operation is performed.

7. The wellbore stability prediction and optimization system based on real-time monitoring data according to claim 6 is characterized in that: The optimization prediction module evaluates the stability of each tetrahedral mesh and performs a mesh density refinement operation or a prediction operation based on the evaluation result, and the process includes: Obtaining scene features of each tetrahedral grid in a three-dimensional geometric model and a formation geometric model in several historical acquisition cycles, performing statistical analysis on load attributes and drilling parameters in the scene features, obtaining the cumulative number of times each type of indicator in the load attributes and drilling parameters is not located in a corresponding threshold interval, and obtaining weight coefficients of each type of indicator in the load attributes and drilling parameters according to the cumulative number; Analyze the load attributes and drilling parameters in the scene features of each tetrahedral mesh in the three-dimensional geometric model and the formation geometric model in the current acquisition cycle, obtain the fluctuation coefficients of various types of indicators in the load attributes and drilling parameters, use the fluctuation coefficients of various types of indicators in the load attributes and drilling parameters of each tetrahedral mesh as evaluation indicators, preset the stability level, and obtain the membership matrix of each tetrahedral mesh to the stability level through fuzzy comprehensive evaluation; The stability level of each tetrahedral mesh is obtained according to the membership matrix and the weight coefficient of each type of indicator. The mesh density is refined for the tetrahedral meshes whose stability level is less than the preset stability level threshold, and then the prediction operation is performed on each tetrahedral mesh.

8. The wellbore stability prediction and optimization system based on real-time monitoring data according to claim 7 is characterized in that: The process of refining the mesh density includes: A mesh density comparison table is constructed, wherein the mesh density comparison table includes mesh densities corresponding to different stability coefficients, the mesh density of the tetrahedral mesh is obtained according to the mesh density comparison table and the stability coefficient of the tetrahedral mesh, and the tetrahedral mesh is divided into a number of new tetrahedral meshes of the same specification according to the mesh density.

9. The wellbore stability prediction and optimization system based on real-time monitoring data according to claim 8, characterized in that: The process of performing prediction operations on each tetrahedral mesh includes: A wellbore prediction model is constructed based on deep learning, and scene features of each tetrahedral grid in a three-dimensional geometric model and a formation geometric model in several historical acquisition cycles are used as a training set and a test set. The training set is input into the wellbore prediction model for training until the loss function training is stable, and the model parameters are saved. The wellbore prediction model is tested by the test set until it meets the preset requirements, and the wellbore prediction model is output; According to the prediction scene features of each tetrahedral mesh output by the wellbore prediction model, the load attributes and drilling parameters in the prediction scene features of each tetrahedral mesh are compared with the corresponding threshold intervals. If a certain type of indicator in the load attribute of the tetrahedral mesh is not located in the corresponding threshold interval or a certain type of indicator in the drilling parameter is not located in the corresponding threshold interval, the prediction scene features are input into the control database for data retrieval, and scene features whose similarity meets the preset standard are obtained in the control database, and the drilling parameters matching the scene features are obtained, and the drilling parameters are used as new drilling parameters of the tetrahedral mesh.

Images